Agreed on the first point, and I think the way to do that is to just not render content unless there's a trust-path to its author. No need for bolt-on ad-blockers, just revoke trust in the advertiser and you'll never see an ad from them ever again.
The only reason there's a connection between the richness of the content, and the likelihood of it being an ad, is because advertisers attract attention by making things shiny. But if we short circuit that upstream, by not running untrusted js to begin with , then there's no reason that we can't have shiny content downstream of where the trust was verified.