Currently we are batching all the requests to Efl.Model without regard for when we need them nor if they are going to block the main loop for to long. The idea is to just wait for the request to only be triggered when the main loop is free with an idler using priority information. For the one that are already in flight, but require additional treatment or more asynchronous request, they should also get a yield added to delay the next step to when the main loop is free. Finally the Efl.Ui.CacheFactory/Efl.Ui.WidgetFactory could also yield its request depending on the main loop free time for this iteration.