Bilingualism (BL) is increasing around the world. Although BL has been shown to have a broad impact—both positive and negative—on language and cognitive functioning, cognitive models and standards are mainly based on monolinguals. If we take cognitive performance of monolinguals as a standard, then the performance of bilinguals might not be accurately estimated. The assessment of cognitive functions is an important part of both the diagnostic process and further treatment in neurological and neuropsychiatric patients. In order to identify the presence or absence of cognitive deficit in bilingual patients, it will be important to determine the positive and/or negative impact of BL properties on measured cognitive performance. However, research of the impact of BL on cognitive performance in neuropsychiatric patients is limited. This article aims to compare the influence of the language (dominant—L1, second—L2) used for assessment of verbal cognitive performance in two cases of bilingual neuropsychiatric patients (English/Czech). Despite the fact that the two cases have different diagnoses, similarities in working memory and verbal learning profiles for L1 and L2 were present in both patients. We expected L1 to have higher performance in all measures when compared with L2. This assumption was partially confirmed. As expected, verbal working memory performance was better when assessed in L1. In contrast, verbal learning showed the same or better performance in L2 when compared with L1. Verbal fluency and immediate recall results were comparable in both languages. In conclusion, the language of administration partially influenced verbal performance of bilingual patients. Whether the language itself influenced low performance in a given language or it was a result of a deficit requires further research. According to our results, we suggest that an assessment in both languages needs to be a component of reasonable cognitive assessment of bilingual patients.