Treatment of the epilepsies have benefitted immensely from study of animal models, most notably in the development of diverse anti-seizure medications in current clinical use. However, available drugs provide only symptomatic relief from seizures and are often ineffective. As a result, a critical need remains for developing improved symptomatic or disease-modifying therapies - or ideally, preventive therapies. Animal models will undoubtedly play a central role in such efforts. To ensure success moving forward, a critical question arises, namely "How does one make laboratory models relevant to our clinical understanding and treatment?" Our answer to this question: It all begins with a detailed understanding of the clinical phenotype one seeks to model. To make our case, we point to two examples - Fragile X syndrome and status epilepticus-induced mesial temporal lobe epilepsy - and examine how development of animal models for these distinct syndromes is based upon observations by astute clinicians and systematic study of the disorder. We conclude that the continuous and effective interaction of skilled clinicians and bench scientists is critical to the optimal design and study of animal models to facilitate insight into the nature of human disorders and enhance likelihood of improved therapies.