<p>In the 3D inversion modeling of gravity and magnetic potential field data, the model weighting function is often applied to overcome the skin effect of inversion results. However, divergence occurs at the the deep area, and artificial weak negative anomalies form around the positive anomalies in the horizontal direction, resulting in a reduction in the overall resolution. To fully utilize the model weighting function, this study constructs a combined model weighting function. First, a new depth weighting function is constructed by adding a regulator into the conventional depth weighting function to overcome the skin effect and inhibit the divergence at the deep area of the inversion results. A horizontal weighting function is then constructed by extracting information from the observation data; this function can suppress the formation of artificial weak anomalies and improve the horizontal resolution of the inversion results. Finally, these two functions are coupled to obtain the combined model weighting function, which can replace the conventional depth weighting function in 3D inversion. It improves the vertical and horizontal resolution of the inversion results without increasing the algorithm complexity and calculation amount, is easy to operate, and adapts to any 3D inversion method. Two model experiments are designed to verify the effectiveness, practicability, and anti-noise of the combined model weighting function. Then the function is applied to the 3D inversion of the measured aeromagnetic data in the Jinchuan area in China. The obtained inversion results are in good agreement with the known geological data.</p>