"“Go looking for your user base” is something that you need to do since day one when you start writing the first line of code" in point 1
Don't you contradict yourself?
Startup is not about creating stuff. It is about solving a problem. People will not be like "oh yeah, I didn't know I had this problem!" So, spending 4-6 month to discover that is extremely expensive.
"Remember that "not-remote" is intrinsically better than "remote" and that brings many advantages." - well, that is a very hard story to sell in 2020
The art is getting people to say “this part of my day sucks” without inviting them to specify the solution. Your job, as founder, is to understand the problem deeply enough, then create a solution, then get feedback and iterate.