Often FO experiments are carried out with the prevalent FO lab-scale setup using a moderate-sized flat sheet membrane. This necessitates spending funds on providing FO apparatuses and using higher area of FO membranes associated with a larger amount of chemicals to evaluate the performance of the FO process in rejection of compounds of interest. In this paper, we used two different FO setups, a typical lab-scale system using flat sheet aquaporin membrane as well as a small pilot-scale setup using hollow fiber aquaporin membrane, to remove three pesticides from water: 2–6 Dichloro-benzamide (BAM), 2-methyl-4-chlorophenoxyaceticacid (MCPA), and 2-(4-Chloro-2-methylphenoxy)propionic acid (MCPP). The obtained rejection data of BAM was compared to our previous study on aquaporin FO membrane using a tiny FO system to investigate the effect of going from a tiny lab-scale to a pilot-scale setup. The diffusion-controlled rejection of BAM was found to be 93–94% using the standard lab-scale flat sheet membrane system while that was 98.7% using the pilot-scale hollow fiber membrane which is in accordance to our preliminary result of >97% with the tiny system. This study confirms the hypothesis that we can employ a very small piece of membrane and accurately use it to estimate the rejection performance of a pilot-scale FO membrane system. This finding allows for a cost-effective and quick test of a large number of molecules by most importantly, the use of a tiny piece of FO membrane. Moreover, both systems, in particular, the hollow fiber setup demonstrated an excellent rejection capability for the charged phenoxy acid herbicides by having rejection values of 99.2% for both MCPA and MCPP.