An extended theoretical model based on a two-scale coupled momentum balance method is proposed to estimate aerodynamic effects of wind turbine towers on the performance of both ideal (infinitely large) and more realistic (large but finite-size) wind farms. A key implication of the extended model is that a normalized support-structure drag, AS/A×CD*, where A and AS are the rotor swept area and support-structure frontal projected area, respectively, and CD* is an effective support-structure drag coefficient, may play an important role in the design of future large wind farms. For the infinitely large case, the theoretical model shows that the optimal turbine spacing should increase with the value of AS/A×CD*, whereas for the large finite case, this also depends on an additional parameter describing the response characteristics of the atmospheric boundary layer to the total farm drag. To validate the theoretical model for the infinitely large case, Wall-Modeled Large-Eddy Simulations of a periodic array of actuator disks with and without support structures are conducted. The results show a reasonably good agreement (within 10% in the prediction of power) with the theoretical model.