No offense, but the subject you are asking about was already discussed in great detail in the very same UbiSoft discussion to which you provided a link.
The only new information you presented in your first post to justify further question or discussion is Raptorman's article. That is, "F4 models it this way, therefore F4 is correct." This is a circular argument that totally ignores all of the real-world research that went into the discussion to which you provided a link, and replaced it with F4 research in the article by Raptorman.
I don't see why I or anyone else should repeat themselves to convince you. The knowledge presented in the discussion that you linked is not based on F4 or Lock On but rather on real documentation, including the recently declassified Russian Gardeniya jammer manual. If you believe Raptorman more than those sources, then you should direct your question to Raptorman, regarding whether he did any research on real-world ECM and ECCM at all, or if rather his article is confined strictly on how to play F4 according to its own rules.
As a counter-example, I also recently posted a question in an F4 forum, asking for an explanation of one of its "as realistic as in real world as possible" features:
http://forums.frugalsworld.com/vbb/showthread.php?t=94591
As you can see, there were no replies. The truth is that even F4AF is no better than Lock On in the radar realism department, and is in many ways worse. F4 just has a larger community of enthusiasts who "want to believe," that's all.
As for discussing accurate ECCM without having an accurate ECM model to react to - that just doesn't make any reasonable sense to me at all.
-SK