Ed,
One thing RCers do that makes a lot of sense - they generally put the on/off switch on the side away from the exhaust.
Almost all profiles have the engine outboard, and our models are 'juicier' than RC models (we use at least some castor in the fuel.) So, the hole for an inboard pushrod is on the "dry" side of the fuse. Should reduce fuel soak into the wing root area. That also goes with the reversed - UP line forward - bellcrank...
Why Up line forward? The prop acts like a gyroscope. Tilt the 'axis' up, and the gyro tries to turn its axis to the direction 90° 'later' in the direction of rotation. For most models, that means that UP input tries to swing the nose outward, and DOWN tries to swing it inward.
The control lines take different loads when we give a control motion - the load shifts toward the side we need to pull against the air loads on the moving surfaces. With UP line forward, that has a tendency to pull the nose in, IOW, to oppose the gyro effect of trying to turn the nose out. And vice versa... So, there's less tendency for the model to change its attitude in yaw (left/right).
Bellcrank position is NOT important for trimming a model to fly well. LEADOUT position is! The pull aims through the leadouts toward the CG, wherever the bellcrank is. Putting the bellcrank in line with the angle the lines make when they reach the tip guides can reduce bending the leadouts, and the friction that comes with that.
It is also important to mount the bellcrank pivot for strength - which usually means near the fuselage centerline and maximum rib depth wing chord location.
To get the UP line forward, you can either turn the bellcrank over from the traditional positioning (pushrod inboard of the pivot), OR put the elevator control horn above the stab/elevator chord line. ...Btw, that puts the flaps horn under the wing chord line.
Done right, a model will not swing in yaw on sharp control inputs, and the controls will remain blissfully free for a long time. This is a very simplified description. I've played with some calculations to try for a complete cancellation of the gyro effects, and for where leadout guides should be with a given CG location. I think it works, for me, anyway...