There’s really only one thing you have to take away from ‘The Mythical Man Month’ and ‘Peopleware’ in my opinion (more people != more work), but they’re still valuable reads.
Man month is also pretty dated by now. There is some valid advice in the book still, but oftentimes it has to be 'translated' to the modern computing environment. Other things in it are just completely outdated.
I wouldn't recommend it as a starting point, but I learned a lot from The Mythical Man Month. It seems like every agile/scrum/kanban/lean book starts by arguing that the old ways of managing projects had problems. It was useful for me to read Brooks' deep dive into what, exactly, those old ways were. Some of the methods stand the test of time and others have not (e.g. not using too many comments in your code because of how much disk space they consume).
DeMarco's Peopleware is a bit more hands on compared to the man month.