Models play an important role in software engineering. On the one hand, we see the broad adoption of model-based approaches for code generation, testing, and other aspects of the software engineering process. On the other hand, models are central tool for conceptualizing the software to be build and are therefore of vital use during requirements engineering, architecture design, and test planning. Although models are widely used in software engineering, and hence prominently taught at university level, we still have little insights into objectively assessing the quality of a human-created model. Whether a model is “good” or “bad” largely depends on the subjective perspective of the person reading the model, and is usually independent of the notion whether or not the model is syntactically or semantically correct, or adequate for its intended purpose. To gain insights into what makes a good model, we have proposed a study to evaluate student-created models using metrics suggested in the literature. In this paper, we provide a comparison of case examples for different model types to highlight initial observations about the usefulness of these metrics to objectively assess the overall quality of a model.