The neuronal ceroid lipofuscinoses (NCLs) constitute a group of rare and fatal autosomal pediatric neurodegenerative disorders. The most prevalent subtype, CLN3, is characterized by a mutation in the CLN3 gene that encodes for the CLN3 protein. No treatment options for this disease exist. Although common pathological hallmarks have been identified, the pathophysiology of CLN3 disease is poorly understood, limiting pharmacological target identification. These hallmarks include the accumulation of lipofuscin and subunit c of mitochondrial ATP synthase, mitochondrial dysfunction, and attenuated Bcl-2 expression. Herein, we report the synthesis and characterization of strategically designed bicyclic analogues of the bioisosteric non-opioid analgesics Flupirtine and Retigabine, previously shown to exhibit neuroprotective effects. Metabolic action on the parent structures leads to the formation of reactive diamine/diimine intermediates that result in adverse effects, including hepatotoxicity. 1H-benzo[d]imidazoles incapable of forming the known toxicophore demonstrate enhanced protective effects in a highly phenotypic CLN3 patient-derived induced pluripotent stem cell (iPSC) model. Selected lead compounds 9b and 38b afford significant protective effects and have a mechanism of action to activate autophagy which results in the capability of rescuing a CLN3 disease hallmark, the accumulation of subunit c of mitochondrial ATP synthase. Furthermore, both compounds possess ‘drug-like’ pharmacokinetics and represent promising compounds for CLN3 drug discovery.