Socionics is indeed like a religion, because it is peseudoscience. Pseudosciences tend to always rely on a small basis of faith that some sort of calculation or generalization is true without much empiricle evidence backing it up. However, this does not mean that proof for socionics can not be supported by objective observations and sensory information, it just can not necessarily said to fundamentally be said to have its basis in those methods.

So, your ESTj was right in saying that it seems like a religion taking what I have said above. On the other hand, socionics should not necessarily categorize as a religion necessarily because of that fact.