The results of nutritional interventions with long-chain, polyunsaturated n-3 fatty acids (PUFAs) with regard to the course of psychiatric illnesses are inconsistent. Poor dietary habits and associated nutritional deficiencies have recently been identified as potential treatment targets. The aim of our study was to compare the dietary habits of patients with schizophrenia and major depression and relate them to clinical symptoms at the time of illness exacerbation or age at onset. The frequency of consumption of selected foods (PUFA-rich or high glycaemic index foods) was determined using a self-report questionnaire. Clinical symptoms were assessed using the Positive and Negative Syndrome Scale (PANSS) and the Hamilton Depression Rating Scale (HDRS) for schizophrenia and major depression, respectively. Four dietary factors and the distribution of patients within tertiles corresponding to infrequent, less frequent or more frequent consumption of each food group were assessed by factor analysis. The differences in dietary habits observed between the two patient groups were of medium or strong effect size. The consumption of savoury snacks and sweetened beverages changed the negative PANSS scores by 8.1%, while psychopharmacotherapy in schizophrenia mainly influenced the positive PANSS and PANSS total scores. In major depression, the consumption of olive oil and PUFA-rich foods changed the age at onset of illness by 6.6%. Dietary habits had no effect on the severity of clinical symptoms. By reducing the frequency of eating unhealthy foods, schizophrenia patients were able to lower their negative PANSS scores. Patients with major depressive disorder were able to delay the onset of clinical symptoms if they consumed olive oil and PUFA-rich foods more frequently.