This post is long and boring, so if anything, just read the very last paragraph. That's what's important about my point.
There is alot of information in Socionics on the way people perceive certain situations and how they think them through. Te uses known facts to argue a point, and Ti uses speculation. A person with dominant Fi acts as a moral policeman within arguments and social events. Ne is very interested in crazy theories and people's inherent abilities. Si is interested in comfort.
But does anyone else feel like this all falls apart when you try to grasp the entire personality and not just how they process information in very limited settings? I find Socionics to focus on two main things: perception and judgment (and thus I would agree with the other thread about the importance of J/P in Socionics.) We perceive things in the environment and make judgments on them. We may look for innovative ideas or just for straight-up details. We may examine our perceptions from a logical standpoint or a moral standpoint.
Sorry, folks, but this is not how people work. Over a century of research in psychology has shown this. What about classical conditioning? Or operant conditioning? Or social cognitive learning? Or cognitive psychology? Or neuroscience? Or, hell, even non-Jung psychoanalysis? If anything, these fields have shown us that people are both quite complex and strangely uniform.
It seems to me that one large factor that is missing in Socionics is a theory of action. It's not enough to process information; one must be able to use it, and that's the important part of personality theory. Relational theory in Socionics does take this into account to some extent, and good for it, but this is not the only place action should apply. Try examining the ENFp and his relations to other people in general: how he initiates interaction, how he reacts to their varied actions, and how he creates, maintains, and destroys his social networks. Research on this stuff does exist, but the explanations of the "why's" of action given by Socionics is quite lacking. The function descriptions focus on things that no one really, actually, truly does or cares about.
Let me give an example of what I think would be a good start in the direction of research I am proposing. Take the INTj: normally described in Socionics as a philosopher and a knight for truth, a person who builds logical systems around innovative ideas. In all actuality, that is about the bulk of how Socionics describes the INTj. Sure, there are the Stratiyevskaya descriptions that list mounds and mounds of information about INTjs, but they only take each function description and stretch it into around ten paragraphs of redundancy. And what's worse is that same functional approach: we know plenty about the details of one type, but little about what this type is really about. What's stopping us from pinning people down to one type based on behavior that is not mutually exclusive amongst differing types?
From a social/behavioral standpoint, the INTj is someone in search of an organization for an environment that constantly threatens his well-being. As is common for all NTs, he treats the world and himself as two separate, distinct things. He tries to be organized and rational, and may only become less so in places where he feels safe and unobligated. The problem with the INTj, though, is that there is nothing definite for him. The life of the INTj is marked by the struggle of trying to make sense out of a disorganized, irrational world. The INTj is closely related to his immediate introverted partners, the ISTj, INFj, and INTp. Only here, the ISTj assumes regularity in some part of his life (usually his experience) and thus is able to be much more directed and organized than the INTj. The INFj is less concerned about the world as a separate, impersonal thing and more concerned about his own relation to the environment and other people. The INTp is quite like the INTj in that he is always trying to escape a dangerous world, but instead of coming at it in an attempt to subdue it by organization, the INTp finds it easier to simply avoid conflict situations and attack life in a very roundabout way.
The above description may or may not be entirely correct, but it shows at least the direction I want to go with personality theory. Note that the description focuses on the type's relation to the environment. I find this to be the single most important thing when analyzing a type. Sure there are elements of classical Socionics intermixed in there, but they only serve as a portion of the description, not the whole thing.
Hmm, I really didn't plan on this being an all-out criticism of Socionics. Really, my goal was to explain that Socionics doesn't seem to apply in many parts of real life. E.g. my two best friends are also my cousins, their types being INTj and INTp. That gives us a group of three pure INTs. One would expect us to be one big geekfest, talking about particle physics to neuroscience to historical figures, the INTj getting agitated with the two INTps and their reliance on Te, the INTps ganging up on the INTj and his Ti. Very, very far from the truth. While we do talk about some intellectual topics, it is mostly just an exchange of information, no arguing at all. What we mostly do is play games and run around like idiots and go to the mall and bullshit at the library and in general act like anyone else would. Another example would be my INFj friend. To describe him in terms of just Fi and Ne would miss so much of his personality, I wouldn't know where to begin on listing it.