So to be clear, your recommendation is to *not* include, e.g., nested lists in long descriptions, but to instead convert complex content that could be displayed as a nested list to a nested list in the body text itself.
I struggle with this a bit. I understand that the visual relations of a flowchart are not accessible to everyone, and that alternative methods of access to this material are necessary (e.g., through a nested list), but I don’t agree that converting a flowchart to a nested list in the body text (thus eliminating the visual element completely) is the best solution for everyone. I think it’s useful to have both of these modes of access in the book—some readers will benefit from the visual relations, others from a complex list, and others from a prose-based long description, for example. I don’t really know how to have all three of those things in the book at once, but to me, having all of these elements embodies a universal design in learning approach, as well as the spirit of the “multiple ways” requirement WCAG success criterion.
Apologies if I’ve misunderstood you, and also if this is better suited to the “discussion” section!