Not even the most brilliant conceptual model would be of any use if no one could understand it. A basic precondition for a models' usefulness in practice is that it is comprehensible. In this keynote, I will give an overview on the vibrant stream of cognitive research on conceptual modeling and summarize empirical evidence on how to develop useful and understandable (visual) conceptual models. More precisely, I will present a systematic summary of empirical and theoretical work on the sources of cognitive effort in comprehending visual conceptual models. Such insights can be a valuable enhancement to any modeling course!