Self-folding origami has the potential to be utilized in novel areas such as self-assembling robotics and shape-morphing structures. Important decisions in the development of such applications include the choice of active material and its placement on the origami model. With proper placement, the error between the actual and target shapes can be minimized along with cost, weight, and power requirements. Through the incorporation of dynamic models of self-folding origami mechanisms into an optimization routine, optimal orientations for magnetically-active material are identified that minimize error to specified target shapes. The dynamic models, created using Adams 2014, are refined by improvements to magnetic material simulation and more accurate joint stiffness characterization. Self-folding dynamic models of the waterbomb base and Shafer's Frog Tongue are optimized, demonstrating the potential use of this process as a design tool for other self-folding origami mechanisms.