Looking at it from a teacher's perspective, the problem with DMG is that it is not clear who it is written for or what it is trying to achieve.
It starts with sort of general advice about running a campaign and world-building, but this is all sort of generalized and aspirational, and kind of mashes up two distinct tasks. This actually reads like something intended for an experienced player who is considering trying to DM but already understands the game very well and has lots of ideas already. The tone is almost philosophical ("what is an RP?").
If I was aiming for a new player, I would really break this section down into concrete, achievable chunks. And the number one teaching tool that the DMG lacks is modelling: the best way to teach something to someone isn't to explain it at length, it's to introduce basic concepts, give them models to study, and get them started on a simple version of the task so that they can learn by doing.
And, in early chapters, I definitely would NOT go anywhere near discussing building an entire campaign world, which is where the DMG heads very quickly, sometimes in broad strokes and sometimes in weirdly granular detail. No! Start by just helping them create a basic adventure setting for one game, giving them a model to work with. You want to get the student doing something independently as quickly as possible. Give them an achievable!
Okay, that's Chapter 1 - you've made an adventure (or can just use this model we've given you, if that suits you better for now).
Oh, key point: after every session, they should ask for constructive feedback from their players! Formative assessment is really useful to help improve!
Chapter 2 - how to run a short Session 0 and get playing. Again, just the basics, with some concrete tasks to work on, like having your players fill out brief character cards, make a mutual list of expectations (this is something almost every decent teacher does at the first class of the year), that kind of thing.
Chapter 3: Expanding your world: now you add some more instruction, a little more detailed, on where you can take the adventure next. Again, provide a model. Note that this is looking kind of like how the first D&D worlds were created back in the 70s - there's a reason for that. It's kind of an intuitive, step-by-step process.
Subsequent chapters: maybe one has to do with world religions. Use that as the basis for an adventure. Include a model. Maybe another one has to do with traveling to another plane. Again, get the student to actually build an adventure, and give a model.
In these subsequent chapters, I would add more of those DMing tips, but I would do it in side bubbles, where relevant to the adventure that is being built.
This is all off the cuff, but that is the sort of structure that you use when you are trying to bring new students up to speed in pretty much any subject: you teach a bit, have them do a thing, teach a bit more detail, have them apply it again, now get them to synthesize into something even more complex, and so on, while providing consistent feedback.
That's what I would do with the first half of the DM's Guide, and I would take it far enough that even experienced DMs would find some expert ideas to consider as the chapters progress. But the second half of the text would really be aimed at experienced DMs - like if you've mastered that first half, here is a ton of stuff that you can use to really personalize your campaign. Some of these would be optional rules, but developed enough that they can be plugged right in and tried out. Others would be story hooks and role-playing ideas. And towards the end I would get really out there, with avant garde ideas for different ways to play the game and build adventures.