Principles of design
When designing an airplane, even if "just for RC," there are several factors that one should consider. I can guess that if you are reading this you are looking for some info on proper motor size, prop size, and maybe pitch. While this information is important, it is probably not the best starting point. These factors really come much later in a design. Ahead of the scratch builder are some other daunting details. Details like airfoil, drag, control surface design, and avoiding catastrophic failure in flight can quickly overwhelm the would be builder. Indeed home made design can be very complicated. And then, one would still have to size that motor... It's no wonder there is such a market for the ARF models.
But then again, consider the basic paper airplane that nearly every school kid has made. The construction material is paper, there is no need for glue, and the time to make takes only a few seconds. Yet for the application, that paper airplane cruises wonderfully. If we learn nothing else from that design process, one should take away the concept of keeping it simple. KISS, or "keep it simple stupid" is a phrase that a good designer should be muttering throughout the entire design process. Simple design leads to reliability, ease of repair, and reproduction of results. It will almost aways cut down on construction time and skill required to build. These factors almost inevitably lead to a lower cost to build.
Making the most from the KISS concept is the profile foamie. Simple airfoils, flat bodies, flat wings. Easy to build, almost guaranteed to fly. This website however is dedicated to building full fuselage planes and to do that, realizing the challenge is only the first step. So aspire big, but learn from the small. Maintain the guaranteeing principles, while striving to enhance the scale and impressiveness of your flying model. Once you learn the concepts, you will know ahead of time whether your plane will fly or not, and those guaranteeing principles.
One of the first things to understand is the four forces that act on an airplane. First is the easiest to understand, the constant force of gravity. Gravity will always pull down on the aircraft at the same rate. To counter gravity, there is lift. Lift is more complex and is generated mainly by the wings, but can also come from other parts of the plane during a flight. As any object moves through a fluid, be it a liquid or a gas, there is resistance. As that object increases in speed that resistance also increases. This resistance is known as the force called drag, and it aways pulls in the opposite direction in which the craft is moving. This leaves the final force, thrust. Thrust is the force that propels the craft. Pretty straight forward, and typically generated by the propeller in RC airplanes. When these forces balance, the state of the airplane becomes static. For instance when thrust equals drag the craft will stay at the speed it is currently at. When lift equals gravity the altitude will no longer change. When drag increases, the plane will slow. When lift increases, the plane will rise. It's also important to note that all of these forces act around the COG of the aircraft.
As you consider your model another very important concept to understand is aerodynamic scalability. Very basically this means that smaller scale models will perform proportionally, smaller, to their bigger full scale counterparts. In fact, generally speaking, a model 1/2 the scale will generate 1/2 the lift and 1/2 the drag of it's full scale counterpart. This concept has been used widely in industry and is what makes most wind tunnel testing possible, as the energy required to build a full scale wind tunnel applicable to the speeds in which real aircraft fly is incredible, although it has been done. Even so, the Langley full scale wind tunnel only produced air speed of up to about 120MPH, at least according to wiki. The result of such expensive requirements have lead to testing only parts of an aircraft or testing of aerodynamic scale models. Realize however, the required airspeed to simulate actual flight is scaled as well as the model gets smaller. Translated to the scale RC world, your smaller model may handle like it's real brother, but it will likely not fly at the same speed. This may not be such a bad thing anyhow. For more information as to exactly why this is, look into compressibility and Reynolds Numbers. Incidentally this also means that there is no need to increase the size of any wing or control surface because the model is smaller the the real thing. The only reason for changing the original design would be to make the aircraft easier to control from the ground. For instance a common practice is to increase the size of ailerons to obtain a more responsive roll rate. others increase the size of the rudder to add directional stability, which may help a tip stall problem. Do also realize, this very same process happens in the full size world as well, note the addition of the fin on the later model P-51 Mustangs and P-47 Thunderbolts.
Design to break
Design your aircraft to be crash friendly. Building out of foam is more then half the battle. But, if there was one thing I learned from flying Parkzones it was the reaction of the battery in a crash. It seems I've managed to eject that thing out of every direction from where it belonged. As a result I try to mount my batteries directly to the sturdiest part of the plan, often the wing. I also put the expensive parts in a location where the battery will likely not fly into on a sudden stop. This saves my batteries, and saves equipment. I've also mounted things that like to break off with strong magnets. Sometimes too strong. I had an SE5A that lasted much longer then it should have due to the break apart design. This kept me flying longer, and increased the learning curve on a poor flying model. The other side of this idea is trying to build an entirely indestructible airframe. I have and have often seen people covering the entire airframe with heavy fiberglass to increase durability beyond what is practically required. If these overbuilt airframes do fly, which mine have not, when thrown to the ground something MUST absorb the energy of the impact. Coating the entire craft will only mean you can not control where the breaks and cracks will happen, and will also make it very difficult to repair. For this reason I tend to make my wings very strong, yet attach them with as little glue as possible. In fact I've also used magnets in this application with great success. The trick in all this is to find out what will survive the typical stresses of normal flight. Remember the minimum is really all that is required for RC. The bad part is there is really only one way to learn this.
Push the limits
Don't be afraid to push the limits on what will actually work. Build things thinner, design to break apart, don't worry so much about the drag some external simple linkages will create, and try strong magnets for different ideas. The flag ship of my current air force is a 45 inch wing span P-47 Thunderbolt. This is by far my best flying plane. Plenty of power and plenty big. The entire airframe is held together with rare earth magnets. I never figured it would hold together, and this past weekend I had a run in with the lines of a stunt kite. I should have never of been flying with enough wind to hold a kite in the air, but the 36 Oz monster was really unaffected. But the string from the kite disagreed with my model in the air and grabbed it out of the sky at about 100 feet. The string latched onto the prop and promptly stalled the fighter. Pulling back on the stick was useless. It went straight into the ground nose first. The prop broke, and the nose is slightly crumpled, beyond that the damage is insignificant. The nose simply popped off at the magnets, the prop cutting a small gash in the wing as it flew by, the wing then popped of the fuse, followed by the canopy. If I had another prop it would have been able to go right back into the air. What more could you ask for from a homemade design?
Last, size that motor and prop.
For early designs look for something that that will provide as close as possible to one ounce of thrust per ounce of weight. Also look for a lower kilovolt rating in the 900KV range as this will offer plenty of torque. That way you can get out of trouble fast, while a high speed will fall second to sheer power. Then just adjust the prop size to match the power you need from the motor.