What do you think of dynamic mics instead?
Given the same pickup pattern and frequency response, I don't think it will make a difference. The root of the problem is the upright bass is simply not loud enough acoustically for a mic to work in a noisy environment. The mic cannot isolate and pull the upright bass's weak signal out of the acoustic mayhem.
Any time you use two pickup mechanisms there are potential phase issues. This will be true whether you use a second contact pickup, or a free air mic. Even some single pickups with two or more elements have this problem. You can find quite a few comments where users cut one of the elements off their Underwood pickup. So using the Ishcell with your existing piezo may work great, or not. Part of making it work can be how you process each signal.
If you explore the Ischell website you will see that it can be configured different ways, including with both a contact pickup and an "aerial" mic.
In my experience a mic adds the most realism in the high frequency range. So it's more about air than body. The worst problem I experience with a mic is bleed from cymbals and feedback from the amp. I had the impression that the rejection off the back and side of the mics was adequate. The cymbal problem related more to the sound bouncing off the upright's curved front panel. This caused an effect similar to flanging. So not only were the cymbals loud, but they sounded harsh and distorted. Rolling the highs off the mic enough to mitigate the problem with cymbals eliminated any benefit the mic provided IMHO.
The feedback problem with the amp is solved by not sending the mic to the amp. Use the contact pickup and only send the mic to FOH IMHO.
If you try one of the sE Electronics mic shields, bigger is better. It's not only about shielding the back and side of the mic from bleed. I believe it's also about creating a sound shadow so less sound bleed is reflected off the front panel of the bass back into the microphone.
Choosing a different contact pickup may help. The Ehrlund and Ischell are not 100% authentic to the sound you get from a mic, but they are way closer. There is no best between these two. The Ehrlund make work better on one bass, while the Ischell works better on the next. From what I have read, the Wilson K4 focuses more on the string sound. The pickup elements are very close to the top edge of the bridge. At the opposite end of the spectrum is the Realist Copperhead which sits between the bridge and the body. This Realist work fairly well for bow, but tends to pick up a bit too much body when you play pizz. This often causes the sound to be more diffuse and boxy, rather than clear and punchy.
Another thing you can try is corrective EQ. The bass, pickup, amp, and cab form a system. The frequency response of this system is not linear. As you turn up the gain, the peaks in the frequency response tend to get more extreme due to the feedback loop between the bass and speaker. But identifying and cutting these peaks, you can make the sound more clear and true. I used to do this with either a parametric or 31 band graphic EQ. A modern produce called the ToneDexter does something similar.
ToneDexter by Audio Sprockets - Studio Sound for Your Live Performance
I should have mentioned the Chuck Israels mic by Xlson. I don't have personal experience, but
@rickwolff has posted a bunch of video and it sounds fabulous and seems to give slightly better isolation than the DPA. I don't think it would be effective on a loud stage, but it's a step in the right direction.
Instead of a mic shield, some sort of gobo system may work, but a gobos will be large and expensive.
RealTraps - MiniGobos
You may make a mic work, but it will most likely require quite a bit of experimentation and also is likely to get expensive rather quickly.
