Can't I combine all my translation, rotation and scaling into one model matrix? And then if I want to translate it further I could just do translate on the previous matrix the next frame?
OR do I have to save each translation, rotate and scale as separate matrices for the model?