I now more see you point. You have a two 90 deg bends that are peripendicular towards each other.
Inserting a splitter or vane can have very detrimental effects on flow distribution.
Without experimentation or CFD, I'd not strap it on a engine. If you test on the engine, I would recommend monitoring the individual cylinders via exhaust gas temp, lambda in each primary or pressure in each runner before the intake port to get any idea on distribution.
Regarding a improvement or a way to fix this, I have to first think about it.
But usually air taking bends is not as bad as water / meth or fuel taking bends.
Airflow though a plenum on a inline 4 engine happens only through one runner at a time. This means that the flow distribution, both pressure and velocity, within the cross section of the central inlet and the runner won't change too much. Say the left half of the inlet pipe flows more than the right half. For the big part, the flow in each runner will be higher in the left half than in the right half, too.
Once you run the engine under power, you use excess fuel anyhow. a small deviation in air flow won?t make much difference. water makes the engine even more tolerant of AFR variations, especially on the lean side of things. if it runs very rich, it just loses power.
Most plenum chambers have a turn in the inlet pipe leading to a non-uniform velocity distribution. The air stream then has to take a 90? bend into the runners from a plenum chamber. The feed stream and its distribution of velocity over its cross section is the same for every runner (exception is high boost levels, here velocity variations also cause significant pressure variations).
if all runners would flow at the same time, things would look vastly different.
These video shows the effect to a degree: The air runs through mostly one runner at the time undefended of the flow mess before that
https://youtu.be/299Y50YWMGo
https://youtu.be/jm0b6udpW1c
https://youtu.be/X3ecq8MNfk0
https://youtu.be/6zUwJDUCZ08