This study investigated the strength of lexical and non-lexical processing among Arabic (L1) English (L3)-speaking children (fourth and fifth grades, N = 532) in two writing systems that vary in terms of transparency. Children were assessed using word reading, phonological and vocabulary measures. In Arabic, the study focused on standard form. Assessing the contribution of phonological and lexical routes to reading accuracy involved the use of structural equation modeling (SEM) analysis. The findings revealed that the contribution of the lexical route was similar in vowelized and unvowelized Arabic orthographies, with an advantage for the latter. Findings indicated that only the phonological route contributed to Arabic pseudoword reading, whereas the lexical route did not. In Arabic word reading, the lexical route's contribution was lower alongside more phonological involvement. However, the phonological route's contribution was lower in deep English orthography alonside higher lexical contribution. The findings support the connectionist model and the varies contribution of the latent variables (lexical and phonological) to reading real Arabic words and English words. They corroborate the view that based on the assumption that this difference in orthographic depth exists, lexical-phonological route involvement varies from one language to another which contributed differently to reading in the two languages. However, they support the traditional dual route model in reading Arabic pseudo word, while separate independent phonological contribution was observed. The findings demonstrate the need for the reconsideration of vowelization as the sole criterion in determining the orthographic transparency/depth continuum of Arabic.