The manuscript focuses on structure-activity relationship studies of CNS-active compounds derived from valproic acid (VPA) that have the potential to become second-generation VPA drugs. Valproic acid is one of the four most widely prescribed antiepileptic drugs (AEDs) and is effective (and regularly approved) in migraine prophylaxis and in the treatment of bipolar disorders. Valproic acid is also currently undergoing clinical trials in cancer patients. Valproic acid is the least potent of the established AEDs and its use is limited by two rare but potentially life-threatening side effects, teratogenicity and hepatotoxicity. Because AEDs treat the symptoms (seizure) and not the cause of epilepsy, epileptic patients need to take AEDs for a long period of time. Consequently, there is a substantial need to develop better and safer AEDs. To become a successful second-generation VPA, the new drug should possess the following characteristics: broad-spectrum antiepileptic activity, better potency than VPA, lack of teratogenicity and hepatotoxicity, and a favorable pharmacokinetic profile compared with VPA including a low potential for drug interactions.