There are two ways that you could get "loops inside a loop":

• A single layer interconnected system with $n$ subsystems having interconnections that generate an "internal" cycle: in this case there will be a loop that is shorter than $n$ and so you could, in principle, run a smaller number of iterations to detect the cycle. But running $n$ iterations will always catch this case, so it is a conservative check.

• A multi-layer interconnected systems with $n$ subsystems, and some of the subsystems might have internal loops: in this case when you call sys._compute_static_io there will be a recursive call to the subsystems that are themselves interconnected systems, and you will catch the cycle in the recursion.

It's possible there is some configuration where this simple approach wouldn't work, but I think it is OK.

I don't have any additional intuition beyond what I already provided above ("The basic intuition is"). I am not aware of a formal mathematical basis or proof, but the basic ideas above could possibly be worked into one.