So, i'm just going through the kopp and other RCS figures.
This is a PAK-FA RCS projection from
It claims that the PAK-FA will likely be -40 dBsm, if the intake blocker is correctly installed and RAM cancels 16 dBsm.
If you compare it with the Kopp figures; if you just eye-ball it without using computer-aids, it looks as though the dBsm is 19 at 1.2 ghz (L band) is about -19. On 8ghz, it's -26, and on 12 ghz, it's -33.
This translates to 0.0126, .0025, and .0005 m^2 RCS.
Using Kopp's figure for the projected RAM, we get -22.5, -33, and -36 for 1.2 ghz, 8 ghz, and 12 ghz.
That's 0.0056, 0.0005, and 0.00025 m^2. Not as good as the F-22's RCS, but close under ideal conditions. Another thing to note is that the PAK-FA, versus the J-20, has a significantly reduced cone of RCS reduction; the J-20 tends towards a 30 degree cone of reduced RCS, while the PAK-FA only hits about 22.5 degrees before its RCS reduction starts to lag.
Another interesting thing to notice is that in the indicated projection, the J-20 has a huge frontal RCS spike right on the nose. Perhaps this is to be canceled by interference patterns? No idea.
Of course, remember both these projections are flawed. They don't consider things like traveling waves or other more difficult to simulate electromagnetic phenomenon.