Among predictors of second language (L2) reading, both first language (L1) reading and L2 listening embody the complexities of comprehension ability in their construct. Their contributions to L2 reading have rarely been examined together, probably because of the different theoretical frameworks in which they are postulated. Therefore, the field lacks evidence regarding how these two predictors concurrently explain variance in L2 reading. Employing two layers of latent-trait analyses, this study investigated L1 reading and L2 listening together with L2 knowledge, which is repeatedly shown to strongly predict L2 reading. Data from 325 L1-Japanese university students were first Rasch-analysed to obtain multiple latent indicators of criterion and predictor variables, which were then submitted to Structural Equation Modelling to evaluate the relative weights of the latent predictors. Over 90 per cent of the L2 reading variance was jointly explained by the predictors, with L2 listening and L2 knowledge accounting for most of the variance. A separate analysis of the higher proficiency subsample resulted in L2 knowledge showing the largest contribution, followed by L2 listening and finally by L1 reading.