Does the venturi effect apply in EDF? Will I get more thrust by reducing the radius with a nozzle put on the back of my fan unit instead of just an open end?

You will get less static thrust, but a higher efflux. Generally speaking this 'feels' like more thrust in flight because the model will have a higher in-flight speed.

Yes.

I usually go with 80% of the FSA. FSA is Fan Sweep Area, which is the O.D. of the impeller minus the diameter of the EDF center hub. So it is a doughnut shaped ring. Calculate that area, then you want the outlet diameter to be 100% of that diameter value for thrust, 90% for a mix of speed and thrust, or 80% for speed and less thrust. Once you go below 90% the amps will start to go up from back pressure. On my old Wattage MiG I did the calculations and found it to be exactly 100% which made sense as that was built around a brushed 400 motor using their 6-blade impeller for maximizing thrust using a weaker brushed motor. I set mine up with 4000Kv brushless motor and took 3 blades off the impeller because I was going for more top end speed (trading off a loss of thrust) but it still had plenty of power out of the shoot. I then did the exhaust nozzle duct mod reducing it down to 80%. I gained approx 8 mph with an increase of 3-4 amps.

As far as the inlet goes you want to shoot for 120% ideally from what I have been told. But that's not hard to do if you choose to perform that mod as well because the opening in front of the EDF unit is a complete round circle, not a ring like inside the EDF unit itself. You also want to radius the leading edges of inlets, no sharp inlets. I haven't bothered doing any inlet mods yet as the performance of the EDFs I have is either good as is (for me) or the exhaust duct mod to 80% has given it enough of a performance boost (again for me and my tastes).

Hope this helps.

Inlet is every thing for ducted fans. A properly designed inlet will gain thrust. 50% fsa does work very well if the rest of the duct is designed correctly. Proper lip radius such as a 2/3 elipse and paying attention to the cross section area works. No sudden sharp corners. The transition from the inlet to the fan face is critical. So many designs just use a tube going up to the face which results in a loss of internal speed and pressure ( you're making the fan operate in a vaccume). Transition fsa has to be maintained over the length of the spinner or use a fixed cone to smoth the air to the fan. 120% fsa is a hell of a lot of drag and shows poor design capability.

Study the ducted fan version of the Violet F-86. The inlet is very small but is very well designed and so the plane flies very well.

I agree with what you are saying as the intake needs to be large enough to supply the required air to the fan throughout the entire speed range (different rpms), but yes, if it is too large then that will increase the airplanes drag coefficient and/or decrease the fans efficiency but to make a boilerplate statement that 120% FAS for the inlet shows poor design quality is a little harsh. Like I said I've never done any inlet mods as I've never felt the necessity but just to throw an example out there the recommended inlet for the Hyperflow EDF unit is 129% FSA, the recommended exhaust area is 99% FSA with a minimum of 90% FSA.