In my view, your phrase "If the verification jobs are located in between the 95% confidence intervals" can be closely correlated with the discussion "Does anyone know any recent papers on Statistical Optimum Estimation Techniques?" on the link https://www.researchgate.net/post/Does_anyone_know_any_recent_papers_on_Statistical_Optimum_Estimation_Techniques . In particular, I want to quote a phrase from this discussion:

"Say that b0 is the unknown optimum. We have a known heuristic solution b1 where b1>=b0. Deterministic bounds would be b1 as the upper bound and a1, say, as the lower bound. In this case we know by certainty that the unknown optimum b0 is in the interval (a1,b1). Statistical bounds also provide a lower bound (or rather confidence limit) s1, say. In this case we don not know by certainty that b0 is in the interval (s1,b1), but with a certain degree of confidence. In the experiments we conducted the confidence level was close to one, 0.997. Hence, the statistical bounds provide an interval almost certainly containing b0. "

If your phrase "... between the 95% confidence intervals", in fact,  closely related to the phrase "... that b0 is in the interval (s1,b1), but with a certain degree of confidence" then my answer would be negative: "This is does not mean that the proxy model can be valid for prediction".

try cross validation methods such as "Leave One Out" or other subdividing strategies of the data.

Not 100% sure if you meant to explore the possible solutions by considering measurement uncertainty while calculating error statistics? I've done some work (see the link) where measured data (flow, sediment, and nutrients) are not taken as deterministic values but given certain ranges. Please let me know if you need more information.

Article The Impact of Considering Uncertainty in Measured Calibratio...