I think a lot of people have a separate approach to personality theory.

Some people axiomatically characterize type as something which doesn't change, and set out to accurately build a type theory and characterize themselves.

Some people don't formulate their theories with this axiom, instead types are "bundles of characteristics" that a person can relate to, the mapping of these characteristics to a type though are considered axiomatically to be consistent and unchanging, therefore if they do not relate to a type it requires the introduction of type change or multiple-type theories.

I think this is the fundamental problem.

I'm relatively skeptical of the first method in all honesty, because if it were possible to create a theory that was so consistent to an individual that it characterized them entirely and was a perfect relation to them, then it would require one of two things. First, it would require a very complex range of types, as their is a large variation in people's psychology from individual to individual. Second, the alternative is that instead of a large number of types, you could have a small number of types, but that would imply little variation from individual to individual.

I think that's why it would be hard to create a comprehensive and complete "type" theory. It would either be incredibly inefficient and complex or it would seem to imply little variation between individuals.