A mental model is a simplified representation of how something works. The danger is treating mental models as conclusions — as final answers about how the world is — rather than as tools for navigating it. Tools can be picked up, put down, evaluated, replaced. Conclusions calcify.

The best thinkers carry a large toolkit of mental models and move between them fluidly. They use first-principles thinking for some problems, analogy for others, inversion for others still. They know that no single model captures everything, which means they're always triangulating rather than resting on one frame.

This is infrastructure thinking applied inward: the models you build don't just help you solve one problem, they shape what problems you're able to see. Building better mental models is not a task you finish. It is a continuous project of upgrading the cognitive substrate on which all your other thinking depends. knowledge-as-network