Dyslexia is a learning disorder of neurobiological origin and it is characterized by an unexpected difficulty when learning to read. Its main obstacle is a difficulty to learn the alphabetical principle and establish an association between graphemes and phonemes. This learning disorder affects readers of any language, for instance, Spanish (shallow orthography) or English (deep orthography). Therefore, the main aim of this paper was to investigate bilingualism (Portuguese-English) as a mediator of reading in individuals with dyslexia. For the Portuguese protocol it was administered a battery of tests of the Brain Institute of RS (project ACERTA) and for the English protocol it was used Woodcock – Johnson Achievement Battery – III (WJ-III). The objectives of the present article are to investigate (1) the relationship between reading performance and bilingualism in dyslexia; (2) the neural correlates regarding the performance of bilingual and monolingual individuals with dyslexia, as well as typical readers; (3) reading and writing performance in bilingual individuals with dyslexia; (4) the neural correlates of shallow and deep orthography of the two languages in bilingual reading. The results showed a superior performance of dyslexic bilinguals (DB) in comparison to dyslexic monolinguals (DM) in all reading and writing components in Portuguese. In the task accuracy and orthographic fluency in Portuguese, the DM group had a mean error of 28% (SD 9.38), however, in the same task, but in English, the group had a mean error of 48% (SD 8.50). In the task accuracy and fluency in the reading of words and pseudowords in Portuguese, the DB group had a mean error of 6% (SD 2.38), however in the same task, but in English, the group obtained an average error of 19 % (SD 2.58). We concluded that the learning of an L2 has positively influenced the DB group of the present study, corroborating the ideas of Sparks and Ganschow (1991), concerning the Linguistic Coding Differences Hypothesis. In the comparison of DB performance in Portuguese and English tasks, the results suggested that the apparent superiority in DB performance in Portuguese tasks may be related to the orthographic depth. The results of the neuroimaging (fMRI) showed that the dyslexics do not present activation in the visual word form area (VWFA) for words, the typical do, though. Dyslexics have more activation for false font. This result shows that the VWFA has not yet automatized and adapted to the identification of the visual form of words, it is still responding more to figures. As for the Control group, activation is immediate. The results of the neural activation in the task in English and Portuguese suggested that the DB still lack automatization in reading.