Hi @lorimer and thanks for sharing these ideas!
I have to admit, one thing that made me a bit confused about the idea is why it makes sense that the NNS itself owns neurons: it seems to be a bit a loop that a governance participant (neuron) is owned by the overall governance that is in turn controlled by the neuron. Maybe a good analogy to a known governance process might help future readers understand this aspect better?
Regardless, I now also see that the D-QUORUM idea could be an interesting experiment to try out a governance form that is closer to representative democracy - neurons vote once in a while which known neurons (representatives) are trusted to then make the decisions for a while.
My main remaining concerns is your suggestions to modify the NNS governance code and add a new proposal for operating the neuron. Even though I fully support experiments and trying out new ways of using and interacting with the NNS, I am worried about adding more complexity to the governance canister. I think overall we tend to underestimate how quickly such special cases lead to more code complexity, which might introduce bugs and make developing and verifying the governance code harder. For example, I continuously notice how hard it is for people to understand and reason about manage neuron proposals (which are proposals that target a neuron management function and are thus a bit similar to what you are looking for).
Therefore, I think it would be great if we could come up with an alternative way to achieve what you are looking for without the need for a new special proposal. Perhaps one could use a custom canister that controls the neuron and encodes some rules when and why the neuron’s maturity etc can be disbursed? Happy to help brainstorming more ideas!