That's not a comparison of external cloud vs. self hosted though. It is a comparison of different operating models with respect to change control. Either one can exist in both external vs. self-hosted infrastructure.
I've been in places where the engineering team plugs in an old PC under a desk somewhere, gives it a public IP address of there's the production server. That's the self-hosted equivalent of the engineering team having full AWS access and any change is a mouse click away.
I've also experienced places where any change does take a week or two of approvals even though it is hosted on public cloud.
There is a time and place for all approaches. What works best with a three person startup putting up a MVP is quite different from what is best in a very large corporation operating in a regulated environment.
Change control and service delivery time-frames are two different things. When I left BOA in 2017, you could get most production changes implemented with only two days of lead time from a formal change control perspective. But, requesting a new Virtual IP on a load-balancer could easily take two weeks, just for every layer of bureaucracy to wet its beak. And it was impossible to request something so basic without an online service request, and then follow-up emails because the standard service offerings left all kinds of details undetermined and no structured way to provide the information.
Most production changes require a second set of eyes, sometimes from a particular team, but it's all "just" code review. You put your change in the team's queue, their oncall engineer reviews it the same day, you land the change and it gets executed automatically. Most have implemented namespacing so that changes that only affect your own team's stuff can be approved within your team.
This is all on owned hardware. The difference is that we're a SWE driven company (corporate IT is off in its own world, run in the more traditional way, but they don't touch engineering's production datacenters). Infrastructure teams provide APIs, not JIRA forms.
I've been in places where the engineering team plugs in an old PC under a desk somewhere, gives it a public IP address of there's the production server. That's the self-hosted equivalent of the engineering team having full AWS access and any change is a mouse click away.
I've also experienced places where any change does take a week or two of approvals even though it is hosted on public cloud.
There is a time and place for all approaches. What works best with a three person startup putting up a MVP is quite different from what is best in a very large corporation operating in a regulated environment.