The American Epilepsy Society suggests that this priority be merged with priority (#1).
It is indeed a high priority to make models in cells and multiple organisms (fly, fish, mouse) that actually replicate the human mechanisms, e.g., gene knockout may not be the same as gene inactivation from a missense variant. All models have REAL human relevance, it is just HOW relevant. Another way to say it might be: Develop cell and animal models predicated on gene variants identified in genetic epilepsies or on well-defined cellular mechanisms in acquired epilepsies i.e., brain injury, brain cancer, neurodegeneration, neuro-infection, or neuro-inflammation.
Note that the phrasing of "treat genetics of epilepsies" is unclear but should likely be phrased as "treat acquired and genetic epilepsies"