That actually doesn't bother me at all. Any test attempting to group people into a finite number of categories will necessarily look need to limit the analysis to a finite number of dimensions. The more dimensions you look at the more groupings you'll end up with. Add those together and a consequence is you want to have fewer dimensions if possible, but not so few that the groupings are overly broad. Maybe 3 or 5 would have been better, but it's clear that thought went into the decision, and that it was obviously not an arbitrary decision.