This paper describes a study of the influence of a dynamically flexible building structure on pressures inside and net pressures on the roof of low-rise buildings with a dominant opening. It is shown that dynamic interaction between the flexible roof and the internal pressure results in a coupled system that is similar to a two-degree-of-freedom mechanical system consisting of two mass-spring-damper systems with excitation forces acting on both the masses. Two resonant modes are present, the natural frequencies of which can readily be obtained from the model. As observed with quasi-static building flexibility, the effect of increased dynamic flexibility is to reduce the first natural frequency as well as the corresponding peak value of the admittance, the latter being the result of increased damping effects. Consequently, it is found that the internal and net roof pressure fluctuations (RMS coefficients) are also reduced with dynamic flexibility. This model has been validated from experiments conducted using a cylindrical model with a leeward end flexible diaphragm, whereby good match between predicted and measured natural frequencies, and trends in peak admittances and RMS responses with flexibility, were obtained. Furthermore, since significant differences exist between internal and net roof pressure responses obtained from the dynamic flexibility model and those obtained from the quasi-static flexibility model, it is concluded that the quasi-static flexibility assumption may not be applicable to dynamically flexible buildings. Additionally, since sensitivity analyses reveal that the responses are sensitive to both the opening loss coefficient and the roof damping ratio, careful estimates should therefore be made to these parameters first, if predictions from such models are to have significance to real buildings.