[tumba]

APL is not so much a model of computation as a new notation for semantics which already exist. The real semantic explanation of APL is set theory and traditional mathematics.

Probably the best deep ground-level explanation of APL is Iverson’s paper “Notation as a tool of thought.” [0]

The bootstrapping explanation you describe sounds a lot like what Paul Graham did in “On Lisp” [1] and in a much more complex fashion, Queinnec in “Lisp in Small Pieces” [2], both highly recommended.

[0] https://www.jsoftware.com/papers/tot.htm

[1] http://www.paulgraham.com/onlisptext.html

[2] https://www.amazon.com/dp/0521545668

[kd0amg]

It takes a fair amount of work on top of "set theory and traditional mathematics" if you want to actually state a semantics for APL.