Do you have experience in statistics?
1. The entire point of a model is to see the forest instead of the trees. If you want to see all the types then you can already do that by looking at the reality BUT YOU LOSE INSIGHTS with that approach. A model trades defintion for insights, trades some of the trees for the forest.
2. You want the fewest possible groups in a model. Less groups means easier generalizations. I would say 16 types is the sweet spot in this theory. 4 Quadras or Clubs loses too much information. 32 (sub)types, demonstrative vs creative, adds information but it's far lower than the amount of info gained when we went from 4-16.
factor_analysis_job_applicants_scree_plot.png
Iris_dendrogram.png
3. This model won't tell you everything about the psyche and it doesn't need too. That's why so many people stack Socionics with Enneagram because both models show different aspects to personality.