The reason I didn't include "writing what people want to read about" explicitly in the list was that it isn't actionable. You can monitor readership and decide if you are happy about your level of readership or want to increase it. You can listen to your friends and decide to make changes to your blog (or not).
But can you really know a priori "what people want to read"? Without feedback how do you know? Aren't you just guessing what people want? And even if you could know, isn't it relative anyway? I know of very few things that "everyone" wants to read, but lots of things that "some" want to read. The challenge for the blogger is finding that "some" and letting them know that your blog exists. That's why knowing your audience and publicizing your blog to that audience is so important.
The one somewhat objective portion of "what people want to read" concerns human factors: no matter what our range of interests, it is hard to read a blog that is poorly written, badly formatted, and strains the eyes. Hence the points about readability and usability. But even here, I think you have to again consider your target audience. As we have seen from the often heated discussions about the PM user interface, what is readable for one person is fluff and distraction for another. So once again, we are back to knowing your audience.
Ultimately, a good blog is less about "writing what people want to read about" and much more about "finding the people that want to read what you want to write about". You aren't likely to stick with the blog long enough nor write regularly enough to ever get an audience, unless the journey is at least as valuable as the goal.