There is a qualitative difference from model to model. I have a collection of hundreds of my own models, some of which are short and elegant, while others would

There is a qualitative difference from model to model. I have a collection of hundreds of my own models, some of which are short and elegant, while others would print out into books easily (if edited down). It is rare indeed to find a non-obvious, useful mental model that can be boiled down to 5-10 words without losing significant leverage. However the _reason_ why they are boiled down to anything at all is due to a communications burden, such as that confronted while compiling mental models for publishing. That the author offers links to find out more is, I think, a reasonable effort at dealing with that problem. Let's say your friend asks you to summarize the value of _Jane Austen's_ works while you drive them home from the bookstore. You have been asked by your friend to build a mental model. How much reduction of your experience with the books is reasonable, and where does it lose leverage as a result of the reduction? Yet you have an opportunity to help someone make a great new discovery. This process can be very frustrating for the builder of the model. So while I appreciate a high-quality critique of a bad example of something, I have to say it seems like those examples are a bit reductionist and possibly unfair given the sheer number of mental models out there for examination, and the reasons for, or conditions under which they were selected for publishing or communication.