Classical benzodiazepines (BZs) are the most widely prescribed drugs acting on the central nervous system (CNS). They exert their therapeutic effects via binding to the BZ-site of GABAA receptors, and allosterically modulating the chloride flux through the ion channel complex. Given the multiple actions of classical BZs, the serious limitations to their usefulness have directed much research into development of novel ligands for the BZ-site with retained therapeutic effectiveness and minimal side effects. From the studies of CNS-active chemical constituents of medicinal herbs, some members of the family of flavonoids were demonstrated to have moderate binding affinities for the BZ-site. In vivo studies revealed that these compounds were mostly partial agonists of GABAA receptors, and only a few flavonoids were shown to possess antagonistic activities. At effective anxiolytic doses, the actions of partial agonistic flavonoids were often not accompanied by sedative and myorelaxant side effects. Based on structure-activity relationship (SAR) studies, incorporation of electronegative groups to the C6 and C3 on the flavone backbone was found to yield significant increases in the binding affinities for the BZ-site. It was also shown that 2-hydroxyl was a critical moiety on flavonoids with regard to BZ-site binding. These have guided the identification of several synthetic flavonoids with high BZ-site binding affinity and in vivo activity, and further quantitative SAR studies resulted in the development of several pharmacophore models. This review attempts to summarize these findings, which has led to the establishment of flavonoids as potential therapeutics for GABAA receptor-mediated disorders.