But it can be reduced. That's why pilots participate in cyclic check and training, flight reviews etc. It's why we have developed checklists and crosschecking etc. All these came through studies and and research leading to redesigning of systems based around human factors. It's the basis for why T.E.M is now a mandatory competency for pilots.
Tell that to the Renmark pilots operating a periodic review under supervision of a CASA FOI.
James Reason himself details the limits of process based safety in his books. In many ways Tony Kerns work takes over from James Reason. But personal responsibility doesn't fit well with a regulators mind set.
This forum is good at being unforgiving of pilots. But this accident had 4 very well qualified pilots with very good recency flying well equipped aircraft. Both had active IFR flight plans. Both were flying consistent with their flight plans. Personally, I cannot point to anything that would give me any comfort that the same thing would not have happened to me.
The ATSB report acknowledges that both aircraft were identified via ADS-B returns received by the AsA system (as opposed for F24 etc). The ATSB preliminary report acknowledges that the AsA system had the information that indicated a traffic conflict (note that I say system, not controller. Its unknown what the controller was presented). ATSB have departed from typical practice by not making any comment on the recorded radio transmissions in its preliminary report, nor presenting any transcripts. Which is curious.
This is going to be a complex report and I'll put money on the ATSB not publishing a final report for 3 years after the accident. But I'm pretty sure that airspace design (ie class E, CATAF and control step location), radio frequency boundaries, radio procedures and the concept of aircraft self separating in IMC are all likely to feature in the final report. These are all systems based issues.