The study of γ‐aminobutyric acid B receptor (GABAB) activation is of great interest for several brain disorders. The search of new GABAB receptor agonists has been carried out by many research groups. As a result, Baclofen has become the prototypical GABAB receptor agonist. However, several attempts have been made to modify its structure to generate derivatives with improved activity. In this work, we carried out a theoretical and computational study for a wide range of GABAB receptor agonists reported in the literature. Molecular docking and QSAR techniques were combined by using the interaction energies of the agonists with the key residues of GABAB receptor, as molecular descriptors for the QSAR construction. The resulting mathematical model suggests that the activity of GABAB receptor agonists is influenced by three factors: their shape and molecular size (PW5 and PJI2), their constitutional features (ELUMO and T(N…O)) and the energy interaction with GABAB receptor (ETRP278). This model was validated by the QUIK, REDUNDANCY and OVERFITTING rules, and its predicted ability was tasted by the QLOO, QASYM, R02 and rm2 rules. Finally, six new compounds are proposed (35–40) with high potential to be used as GABAB receptor agonists.