X-ray baggage inspection is essential to ensure transport and border security, as it prevents hazardous objects from entering secure areas. Currently, deep learning is the state-of-the-art approach for automated threat object detection and classification. Proper training of these networks requires substantial data; however, the number of publicly available datasets of X-ray images is limited. To overcome this problem, we propose a method for generating new data by superimposing simulated X-ray images of 3D models onto real baggage X-rays, allowing researchers to train deep neural networks without requiring additional imaging or manual labeling. To validate our proposal, we ran experiments using 3D models of wrenches and the SIXray baggage dataset. The results prove that superimposing synthetic threat objects over a real training subset improves detection performance, with average precision (AP) increasing from 90.2% to 93.7%. As modern object detectors process images in real-time, they prove themselves as a feasible approach for aiding inspectors and even fully automating baggage inspection. Moreover, the novel superimposition and colorization techniques presented in this study can be employed in other areas of X-ray imaging.