Down-force increases exponentially with speed. So the faster you go the smaller the spoiler you need to run to generate the same down-force. Now you will need more grip or down-force in order to corner at those faster speeds.
In my personal experience Iv'e felt how body design and spoilers play a large effect in car handling. I ran a Protoform R9 and the car would constantly whip its rear end around and have far too much steering. I switched to a Mazdaspeed 6 and with the same setup the car was instantly on rails. On a very large outdoor asphalt track I had my rear wing break off one mounting pylon and become inline with the car. Even at the "slower" stock 17.5 speeds it sent the rear end all over the place causing me to lose a couple seconds per lap.
As far as actually testing RC cars in real wind tunnels it would be incredibly difficult to get perfect airflow over the length of the car. Most universities only have wind tunnels hardly large enough for an RC car to sit in(and they take up large buildings)- so homemade ones probably can't yield accurate numbers, if you could even measure them. But I would love to see it done/make one myself!
If you were to actually create a wind tunnel, its best to pull the air over the car. I.e. you have a fan behind the car sucking a vacuum over the car. Its more accurate and precise than pushing air over the car (fan blowing onto it).
If you really want concrete numbers I'd say your best bet would be to create the bodies in Solidworks, autoCAD, NX etc and analyze them through the computers. While this only yields theoretical data it would probably be the best way to compare multiple bodies equally.