Exposure to organophosphorus compounds, such as soman (GD), cause widespread toxic effects, sustained status epilepticus, neuropathology, and death. The A1 adenosine receptor agonist N-bicyclo-(2.2.1)-hept-2-yl-5′-chloro-5′-deoxyadenosine (ENBA), when given 1 min after GD exposure, provides neuroprotection and prevents behavioral impairments. Here, we tested the ability of ENBA at delayed treatment times to improve behavioral outcomes via a two-way active avoidance task in two male animal models, each consisting of saline and GD exposure groups. In a rat model, animals received medical treatments (atropine sulfate [A], 2-PAM [P], and midazolam [MDZ]) or AP + MDZ + ENBA at 15 or 30 min after seizure onset and were subjected to behavioral testing for up to 14 days. In a human acetylcholinesterase knock-in serum carboxylesterase knock-out mouse model, animals received AP, AP + MDZ, AP + ENBA, or AP + MDZ + ENBA at 15 min post seizure onset and were subjected to the behavioral task on days 7 and 14. In rats, the GD/AP + MDZ + ENBA group recovered to saline-exposed avoidance levels while the GD/AP + MDZ group did not. In mice, in comparison with GD/AP + MDZ group, the GD/AP + MDZ + ENBA showed decreases in escape latency, response latency, and pre-session crossings, as well as increases in avoidances. In both models, only ENBA-treated groups showed control level inter-trial interval crossings by day 14. Our findings suggest that ENBA, alone and as an adjunct to medical treatments, can improve behavioral and cognitive outcomes when given at delayed time points after GD intoxication.