The current semi-active or even active control strategies have been developed to address a few drawbacks, such as unwanted large displacements created at the base level and system deficiency in adaptation to different types of seismic excitations, in the base isolation systems. In this article, two control strategies, multi-objective modified clipped optimal and adaptive fractional order fuzzy proportional–integral–derivative, are proposed for semi-active control of a smart base-isolated structure equipped with a magnetorheological damper. The main objective is to reduce the displacement of isolation system without allowing significant increase in the acceleration of superstructure for both far-field and near-field earthquake excitations. Using proper fitness functions, the weighting matrices of the multi-objective modified clipped optimal controller are tuned using multi-objective optimization. Then, the parameters of the fractional order fuzzy proportional–integral–derivative controller are obtained. Next, the fuzzy rule weights of the fractional order fuzzy proportional–integral–derivative controller are updated online based on the values of ground motion and structural responses using an adaptive strategy. For comparison, two control cases in which the magnetorheological damper is in passive mode, passive-off and passive-on, are considered. Numerical simulations show that the proposed adaptive fractional order fuzzy proportional–integral–derivative controller better mitigates the seismic responses of a base-isolated structure excited by a range of real-data earthquakes.