If no universal standard can be applied to judge between sets of societal predispositions (which is to say between societies), then no society can be usefully judged except in its own terms; this is the view known as moral relativism. Taken to its logical extreme, the reductio ad absurdum, moral relativism carries all the flaws it is accused of; murder is indeed equivalent to jaywalking, to a hypothetical detached intellect judging the two.
However, there are at least three standards which can be applied across societies which are useful for comparing them irrespective of the specific tenets of the compared. Our hypothetical detached intellect may still, without moral judgment, scale any two given societies (sets of interacting presuppositions) on the bases of internal consistency, of adaptability to change both external and internal, and ultimately of utility in terms of survival.
The former may be judged by the degree to which an individual member of a society can think and act within the society’s precepts without conflicting with others doing likewise. A society which values only adherence to a given set of self-consistent values would always score highly on such a scale, and need never fear destruction from within; destruction from without would always be a danger.
Most religions score pretty badly on internal consistency; this is why they either split incessantly into smaller schisms, or actively suppress societal mechanisms (such as widespread literacy) for examining their precepts. The latter tactic badly degrades societal adaptability.
Contrastingly, science by its core methodology incorporates any novelties of data or ideas into its core. Basic inconsistencies, such as the conflicts between general relativity and quantum mechanics, are not suppressed or ignored but rather actively worked upon. Nothing is sacred except the methodology itself, and if you could empirically convince scientist that tossing chicken bones worked better they’d probably order all further lab lunches from KFC.
Adaptability is thus built right into the foundation of science. Contrarily, since religion relies strongly upon revelation for content, and such revelation is normally decreed unquestionable (one might almost define religion in such terms), the bases of any given religion are by definition unchangeable and thereby inapt for adaptation either internal or external.
The remaining criterion by which our hypothetical detached intellect might judge societies is simple ability to survive. The Darwinian viewpoint prefers the society that survives best, but there is no inherent guarantee that the society which survives best prefers the Darwinian viewpoint. It could well be, as you imply, that some particular set of non-true societal assumptions might work better than the truth, and even that some non-scientific societal processes might better serve any given good, including survival. With respect, though, I consider it very unlikely; I have to confess, of course, that what makes me consider it unlikely is largely the application of the scientific method itself.
While the body of societies which have appreciated the scientific method (of testing ideas against reality) is necessarily small compared to those which had no such notion, the former have been far more successful, not only in traditional terms of population and wealth but in terms of adaptability to changing circumstance.
Without knowing it for an inconcontestible fact, I therefore believe that a secular society based on evidence-based inquiry is ‘better’ than a religious society based on authority-based faith., where ‘better’ is based on societal survival. Is it possible that a society based on unfounded assumptions and unsound (or no) methods for testing them would be better suited for survival than a society which valued truth and logic?
Well, possibly, but I’m not betting on it. Show me a study, or a disprovable theory.