I think we're getting a lot closer to the meat of the issue here. I agree with you that it is possible to delegate and distribute /most/ of what I mentioned successfully to varying degrees. But that leaves the "incompressible" part of the CEO, which is fundraising/investor relations and hiring/org building.
Fundraising is completely impossible to delegate (if you delegate that to someone else, you'll alienate investors) and even with hiring/org building, you have to at least do the recruiting/candidate selection of top lieutenants yourself as a CEO in order to make sure that you are putting the right team of divisional heads in place to execute. You have to be a better coach than the vast majority of coaches out there to succeed in this endeavor, and to even attract and accurately assess the kind of talent that will set up the org for success. It is very easy to mess this up.
On the whole, a lot of what I'm talking about is high stakes high rewards strategic dealmaking. This isn't a common skill for most people to have. It's a blend of analytical (determining leverage points, BATNA in negotiations) and creative (figuring out how to grow the pie) that is hard to practice unless you've done it before.
Most decisions you make here are irreversible. The other challenge with this kind of dealmaking is that the stakes are so high and the counterparties so formidable that if you put anyone except an equally formidable party in place to negotiate with them, you'd end up in a bloodbath that ended up unfavorably for the company and favorably for the counterparty.
In my opinion, that's why we end up at the steady state of executive compensation across most of the world that we see today. It's a miserable job unless you really enjoy it because everyone will (and should) blame you when things go wrong. That is made more challenging by the fact that by the time one becomes a successful CEO, one can generally retire off earnings and/or become an investor with much better WLB and risk adjusted returns.
PS: I'm also realizing another interesting petri dish where we could see parts of what you're suggesting play out are DAOs in crypto. Theoretically, if we can arrive at a DAO that works efficiently, it would serve as proof that for certain kinds of commercial organizations, the chief executive role may be able to be diminished or even eliminated. I am very curious about whether these experiments pan out (and inside my heart hope they do) although I remain skeptical.