LMM Failure: Mazda Performance Issues

Good morning.
Thank you again for all your help with my previous thread.
Current problem: increased fuel consumption, lack of power and torque. On a dynamometer test, the vehicle produced approximately 73 kW instead of the expected 81 kW. This is a Mazda 626 diesel with direct injection, manufactured in December 2001.
Okay, so I just unplugged the mass airflow sensor (MAF sensor), and here's what happened: the turbo lag in first gear seems to have improved rather than gotten worse, and otherwise, the engine is running unchanged.
Isn't that proof that the LMM is the problem?
Why does the vehicle perform better, rather than worse, after I disconnected the MAF sensor connector? Also, the engine easily reaches the rev limiter.
I don't understand the function of the LMM right now.
Conclusion: If the engine performs no worse with the MAF sensor disconnected than it does when it's connected, then the MAF sensor is likely the cause of the power loss and increased fuel consumption, right?
What do you think about that?
I was only surprised that the Mazda's error memory didn't register any errors.
Furthermore, I once had a faulty mass airflow sensor (MAF) in a VW TDI, and it was very noticeable when the vehicle was running in limp mode.
I hope to receive many responses and wish you a sunny Sunday.
Robs.

As I mentioned in my previous post, there is no error entry for incorrect LMM (mass airflow sensor) values. I believe that the control unit continues to use these incorrect values to determine the fuel injection amount as long as the LMM is still providing readings. If you disconnect the connector, the engine will run in a limp-home mode. This means that the control unit uses fixed values (likely a conservative curve) for the air mass and determines the fuel injection amount based on these values. The values used in this mode may be higher than those provided by a faulty LMM, which could potentially lead to the engine running better. However, to be sure, you need to measure the LMM values while the engine is running and compare them to the expected values.
However, if the controller is running in emergency mode, it will always lose power!

Best regards, Johannes.

Many thanks to Johannes and all the other participants.
I will have the new IMM installed and then let you know...
It's frustrating that Mazda itself dismissed this as a possibility back then, those so-called 'experts.' The original statement was: 'It manifests itself differently than through a reduced performance and increased fuel consumption...'
Have a good day and a nice Sunday.