[MENTION=14825]bellisima[/MENTION]: there are a lot of things I wish people who are reasonably unbiased told me when I first started learning this stuff.
Here's some of those things:
- the MBTI test is constructed using similar methods to the Big Five dimensional theory of personality. In this view, it's totally reasonable to suggest someone is more an X than a F or a T, meaning they are more in between than one or the other.
- The functions theory is very different from the dimensional model (that is, the model the actual test is based on). People very frequently find a very low correspondence between the test and any kind of functions model,despite the apparent translations like INTJ=NiTe, ISTP = TiSe and so on.
- as far as achieving consensus, the MBTI test is a lot more empirical than any functions theory; the Big Five is the most widely recognized personality theory in academic psychology, and impressively, the MBTI dimensions line up well with it.
- People rarely realize how much the functions theories out there have conflicting models, because they just hear of the MBTI and think the model most popularly associated to it is somehow objective or with some kind of consensus. E.g. in Jung's theory, it is perfectly valid to have a stacking like NiTiFeSe, whereas in the model pioneered by Harold Grant, the only acceptable NiT type is a NiTeFi type. Other prominent theories -- Myers started off believing in either agnosticism on the attitude of the tertiary or even possibly claiming it the same as the aux (i.e. either NiTeF or NiTeFe!!). Socionics claims people's conscious functioning is all in the static or dynamic attitude, and their interpretation of some version of Grant's stacking is to say said stacking constitutes the valued information, not necessarily the conscious information; Beebe suggests that the opposing attitudes to the 4 Grant function-attitudes in a type constitute the shadow (in contradiction to Jung's view that the inferior 2 or 1 functions constitute the shadow region of one's function-type). Singer and Loomis, 2 Jungian analysts, decided to operationalize Jung's function-attitudes in a test, where they didn't find any established pattern of identification with the function-attitudes. More or less there's nothing incredibly objective about one functions-theoretic model over the other, and it seems it's one's matter of philosophical taste which appeals better.
I recommend treating functions theories as experimental ideas which you can gain insight from with a flexible view, but not as things you should use in the same conclusive spirit you can use the more soundly empirically validated theories (although among those, some do some things better and others do other things better, naturally).
I have often found functions theories more conceptually enlightening than good for realistic practical typing.
Still, I've formed my best-estimate of my own type in my preferred interpretation of functions theories (ILE is socionics notation for NeTi--although I like the structure of the socionics model, I object to a LOT of socionics, so I kind of just use the notation ... realistically my views are more or less my own). But I don't claim it with any kind of rigidity, only that it is my favorite description of myself using functions theoretic ideas.
If any of this is confusing, I am more than happy to discuss it at length.