The bit about unknown knowns does not make sense to me, also it seems to change the point of reference. It is referencing things from third party perspectives whereas the original ones are referencing from first party.

That is, you are talking other things other entities might not know about me that I know myself which makes them, from my point of reference, known knowns. And in the event that I don't know those things myself they just become unknown unknowns.

If you don't change goalposts and keep the point of reference constant, unknown knowns wouldn't exist.

Personally, if I actually had to define unknown known from the individual entity's point of reference it would most probably refer to derived or calculated values. For example, do I know the product of 171 and 342.45673. Do I know I could calculate it? Yes. But do I know it by head? No. Which means it is eventually a known but right now it's still unknown. But even that is a stretch of a definition.

I am convinced Rumsfeld's initial proposition covers the entire space of knowledge. Maybe you could give actual examples of unknown knowns from the individual's perspective.