Comprehension is the ultimate goal of reading, but it is a very complex task consisting of multiple component skills. A number of studies have tested the simple view of reading (SVR; Gough & Tunmer, 1986) in opaque languages, but few investigations of the SVR components have been conducted on transparent languages. In the present study, we tested the SVR model in a sample of 1895 Italian children attending primary school, from first to fifth grade. An assessment battery was used, which included five different tasks: word and non-word reading, passage reading, reading comprehension, and oral comprehension. Hybrid models combining confirmatory factor analysis with path analysis were run separately for each grade. Results indicated that oral comprehension was always the best predictor of reading comprehension, whereas reading accuracy played a significant but minor role. Implications for research and practice are discussed.