The Reason model shown above is not easy to understand since the justification for the number of holes in the cheese layers is not given.
I am happy to see multiple layers (barriers) but which ones to concentrate the corrective action on?
I submit that one or two of these layers probably have far more pass-thru holes than the other layers. Not only that but there is a possibility in this case that even though the layers are defined for ownership, that a failure in one layer may conditionally mean a failure in another defined layer.