Thank you. The 'supermarket' impression is an artifact of that particular quotation, which is from a longer paper. My emphasis was elsewhere.
I wrote (edited, paraphrased):

'A religion, I wrote, is often rationalized by beliefs that articulate an account of the reality that underlies the universe, according to which there is the possibility of fit. That is, the beliefs entail that there is a relation in which a person can but needn't stand to the rest of what is, which is fundamentally appropriate to the way the reality that underlies the universe is, such that standing in this relation is, in and of itself, the greatest human good.'

Holiness and God-realization are examples. I then argued that the religious practices typically aren't merely productive of fit. To perform the practice for the right reason
is ipso facto to enter somewhat into the relationship that is the sumum bonum. I called this relation 'constitutive of fit.'

            I wrote:

'Similarly, taking the Mass may produce a deeper relationship with God, but the Mass isn't a method or technique for getting closer to God. To participate in the Mass is already to stand in the appropriate relation to God. At the same time, taking the Mass may not be essential for fit: we might allow that a man on a desert island who finds a copy of the Gospels might be saved simply by his love of Christ. Nor is fit reducible to the whole set of behaviours we do because we accept a Christian religious world view, for there may be degrees of fit attainable only by God's grace. The force of the claim that a practice constitutes fit is that to perform it is to enter significantly (but not necessarily completely) into the appropriate relation. What matters is that in performing it one ipso facto participates in the relation that is the summum bonum: the practice isn't merely instrumental to the attainment of fit.'

I added: 'This theory explains why religion is nearly as ubiquitous as human culture and why religion is attractive even to those who are not afraid of death or particularly hungry for security.  People generally want to know the point of their existence and where they stand in the scheme of things.  The capacity for metaphysical yearning is perhaps the most human attribute of our species.  A religion explicates a metaphysical connection that (it maintains) is the main point of human existence, and it enables the believer to enter into that relation through the performance of constitutive practice.  ...  Our theory, therefore, explains why so many people care to be religious.'
 
Initially I thought Fit was the whole deal, but Phil Quinn persuaded me that I had to take into account  entirely 'mercenary' religions, even if such are merely possible. Which is why the
paper is entitled 'A Theory of Religion Revised.'

I expressed the revised defintiion as follows:  

' A religion is a system of practices rationalized by beliefs according to which (i) the practices constitute fit, or (ii) they produce a relation to a SR which is of value because it is instrumental to attaining human goods.'

It's meant to be equivalent to the one I gave above:

'A religion is a system of practices rationalized by beliefs
according to which the practices place the practitioner in a
relation-of-value to a SR.:'

I maintain this is what all and only religions share in virtue of which they are religions.