Sounds like you'll need to gear up a couple of teeth to start with then if the temps start to fall then keep gearing up untill they start to rise again.It's a good idea to use a I.R temp gun to make your testing less subjective.Also i've noticed that different grip levels affect motor temps massively.If the wheels are spinning the motor's not really under load and runs cooler so try to use the same surface and keep tyres,slipper settings and driving style the same when your testing different gearings.(Basically keep as many variables the same as you can so you can say that the difference in temp is more likely due to gearing)

I'd get 19 and 21t pinions and a temp gun and spend an afternoon testing different gearings, timing each run so your measuring temps after the same time (and under the same conditions) for each gearing.

This 'real word' testing is a better method than the theoretical ideal gearing.

I'm thinking of getting a 4200 for my MF2 so i'd be interested in your findings.:D

48 pitch gears are one of the easiest to experiment with since there seems to be a multitude of pinions available from many manufacturers.
My personal favorite is the two 6-packs of pinions from Robinson. I'm sure Mike can get them. The pack with odd numbers 15-25 is Robinson part number 1055 and the pack with even numbers 16-26 is Robinson part number 1050. This gives you 12 metal 1/8" bore 48-pitch pinion gears for less than $30. That's a huge range of testing potential for short cash, especially if you have multiple vehicles to use them in.