Just a suggestion: put the original heat exchanger and the system-level hex in a test harness, like this one: https://www.mathworks.com/help/hydro/ug/condenser-evaporator.html. Use a typical condition from the original model, preferrably steady-state, as the boundary conditions for the reservoirs and flow sources. Then try to match the performance of the two heat exchangers in this setup, before dropping it into the original model.
My experience is that it's definitely possible to create an equivalent model using a system-level block for the geometry-based one. Using the Parameter Estimator App, you can even match the performance for a wide range of operating conditions, so they are REALLY equivalent.
