They looked at ways for people to interact more post-lockdown while still keeping the virus under control.
The first scenario involved limiting contact to those people who live in your immediate neighbourhood, the second looked at limiting contact to people you see on a regular basis, and the third involved creating a social bubble with a certain number of people from other households, who then do not go on to form further bubbles.
All three strategies were effective in keeping the spread of coronavirus low, although social bubbles were in some respects more beneficial, experts said.
In an interview with the PA news agency, lead author Dr Per Block said that within each of the three models, people could treat each other as if they were in the same household, and did not have to stay two metres apart at all times.
But it was important that outside of these scenarios, social distancing was kept up, such as when people went to the supermarket, he said.
"Under the first scenario, what could be done is that you meet people who live within your neighbourhood, so you could extend the radius of your contact to a block or two away from your home," he said.
"In the second, the idea is you ask yourself - who are the people you interact with regularly?
"So you might have a group of friends, or you have a family that includes your parents, your siblings your nieces and nephews, and you try to limit interactions to these groups.
"What that means is that you don't meet haphazard contacts - such as that person you see once in a while only for specific activities, you don't meet blind dates or Tinder dates or people that have not been embedded within your (immediate) community.
"The third scenario is very similar to what's been talked about as social bubbles, which basically is keep sticking to the same people.
"So you need to decide on those two, three, or however many people and stick to meeting them."
He said that, in practical terms, it may be necessary to limit the number of people within the bubble.
The larger the bubble, the more "risky it is that somebody would meet somebody from outside", he said.
Dr Block said that two of the scenarios could potentially be combined, such as a social bubble plus seeing neighbours, but that the frequency of interactions would then need to be reduced to limit transmission risk.
He said real-world data was needed to see how people interacted within each scenario, but added: "We can say with reasonable confidence that all of these different options on how to restructure contact seem to work."
Dr Block said each scenario required that "people stick to it, that people understand that it's useful, and that people trust that others will also do it".
"In a sense, it's a good question of solidarity that we are all in this together and therefore we should all stick to the rules," he added.
In terms of transmission risk, the "best scenario" was for everybody to continue to stay at home, he said.
"But of course, this is where we have the biggest psychological and social and economic costs," he went on.
"Now if we would open up society completely... in terms of transmission rates, this would be a disaster.
"So what we have tried to do is go somewhere in between and say, 'how about if we only try to keep our contacts to a minimum, but also try to be smart about who we meet with and structure our interactions strategically?'."
The study was published in the journal Nature Human Behaviour.
The Scientific Advisory Group for Emergencies (Sage) have previously discussed the concept of social bubbles.
The latest Sage minutes available, from May 7, show that while advisers agree there are benefits from social bubbles for wellbeing and mental health, there are risks if they were to be introduced alongside other changes, or if there is poor adherence.
The minutes say: "The effects of bubbles are complex. Introducing bubbles alongside other changes could reconstruct excessive networks, particularly when combined with any increase in contacts in other settings.
"These networks could enable transmission through the population. It will be difficult to assess the effects of individual policy changes on R if multiple changes are introduced together."
Of the latest Oxford study, Linda Bauld, professor of public health at the University of Edinburgh, said: "We need to recognise that lockdown release is about harm reduction, not complete safety."
But she added: "One of the most interesting elements of this research is that it directly addresses a key element of current public health advice in the UK.
"This is the 'stay two metres away from anyone not in your household' rule.
"For separated couples and single people, including our young people, this rule can be perceived as unfair and is unlikely to be followed in the long-term.
"One approach supported by this research is the concept of a social bubble allowing individuals to have repeated close contact with a small group or 'micro-community'.
"This would be appealing for couples who don't live together, or as the researchers point out, a group of carers looking after vulnerable adults, or might even allow those in the shielded category to meet up.
"It was also the most effective strategy included in the research in terms of allowing more contact between individuals while slowing the spread of the virus."
She stressed that human behaviour is not, however, predictable and "won't necessarily mirror what the statistical models in this study predict."