Recently, many efforts have been made to develop N-methyl-d-aspartic acid receptor antagonists for treating different pathological conditions such as thrombo-embolic stroke, traumatic head injury, Huntington’s, Parkinson’s, and Alzheimer’s diseases). However, as side-effects limit the use of most antagonists, new drugs are still required. In this work, we performed a (quantitative) structure-activity relationship analysis of 17 phenyl-amidine derivatives (1a–1q), reported as N-methyl-d-aspartic acid receptor antagonists, and used this data to rationally design the triazolyl-amidines. The best (quantitative) structure-activity relationship model constructed by multiple linear regression analysis presented high data fitting (R = 0.914) was able to explain 83.6% of the biological data variance (R2 = 0.836), presented a satisfactory internal predictive ability (Q2 = 0.609) and contained the descriptors (EHOMO, Ovality and cLogP). Our assays confirmed that glutamate promotes an extensive cell death in avian neurons (77%) and 2a and 2b protected the neurons from the glutamate effect (from 77% to 27% and 45%, respectively). The results of neurotoxicity and cytotoxicity on Vero cells suggested the favorable profile of 2a and 2b. Also, the molecular modeling used to predict the activity, the interaction with the receptor and the pharmacokinetic and toxicity of the triazolyl-amidines pointed them as a promising class for further exploration as N-methyl-d-aspartic acid receptor antagonists.