Enzyme-linked immunosorbent assays (ELISA) for the detection of soluble epoxide hydrolase (sEH), a key enzyme in the metabolism of fatty acids and a biomarker, may increasingly represent an important diagnostic tool. However, there is a lack of ELISAs for mouse sEH quantification, thus resulting in a bottleneck in understanding the pathogenesis of many diseases related to sEH based on mouse models. In this work, nanobodies recognizing mouse sEH were obtained through rebiopanning against mouse sEH in the previous phage display library of human sEH. Later, we developed four ELISAs involving a combination of anti-mouse sEH polyclonal antibodies (pAbs) and nanobodies. It was found that the double antibodies worked as dual filters and had a huge impact on both the sensitivity and selectivity of sandwich immunoassays. The switch from anti-human sEH pAbs to anti-mouse sEH pAbs led to over a 100-fold increase in the sensitivity and a dramatic decrease of the limit of detection to a picogram per milliliter range in format B (pAb/biotin-VHH/streptavidin-poly-horseradish peroxidase). Moreover, we found that the four sandwich ELISAs might demonstrate excellent selectivities to mouse sEH, despite the antibodies alone showing significant cross-reactivity to the matrix, indicating the enhanced selectivity of double antibodies as dual filters. Eventually, for the first time, the ELISA (format B) was successfully used to measure the mouse sEH level in cancer cells with ultralow abundances. The ELISAs proposed here represent a sensitive tool for tracking sEH in various biological processes and also provide deep insights into developing sandwich immunoassays against various targets in terms of both the sensitivity and selectivity.