Beneficial effects of estrogens in the central nervous system (CNS) results from the synergistic combination of their well-orchestrated genomic and non-genomic actions, making them potential broad-spectrum neurotherapeutic agents. However, owing to unwanted peripheral hormonal burdens by any currently known non-invasive drug administrations, the development of estrogens as safe pharmacotherapeutic modalities cannot be realized until they are confined specifically and selectively to the site of action. We have developed small-molecule bioprecursor prodrugs carrying the para-quinol scaffold on the steroidal A-ring that are preferentially metabolized in the CNS to the corresponding estrogens. Here, we give an overview of our discovery of these prodrugs. Selected examples are shown to illustrate that, independently of the route of administrations and duration of treatments, these agents produce high concentration of estrogens only in the CNS without peripheral hormonal liability. 10β,17β-Dihydroxyestra-1,4-dien-3-one (DHED) has been the best-studied representative of this novel type of prodrugs for brain and retina health. Specific applications in preclinical animal models of centrally-regulated and estrogen-responsive human diseases, including neurodegeneration, menopausal symptoms, cognitive decline and depression, are discussed to demonstrate the translational potential of our prodrug approach for CNS-selective and gender-independent estrogen therapy with inherent therapeutic safety.