as to point 1.: For my observation it is pretty clear, that free trade in most cases contributed positively to four dimensions of food security - for most people in the world. But it didn't reach the rural sites at the periphery specifically and came at high - one may say too high - environmental costs. So this appoach needs correction.
As for two (coherence): for 800 million hungry people the current system doesn't work. The trade system is one component in it. Around 70% of the hungry live in rural areas. The theory is, that social safety nets would take care fo the loosers of competitive market economy. But the social safety nets are, if at all existent, translucent. Given the right to food, more coherence would be necessary. For more coherence one approach could be the Rights Based Apporach, but, as we see, it would need more translation into specific agreements concerning the food system. I think, to gain more coherence we must understand the specificity of the food system more in depth.
As for three (compatibility): Because agriculture (the food system) is special, there has been and is in the WTO the special Agreement on Agriculture (AoA). Although the specificity of this sector is addresses with the AoA, the content of todays AoA - see the preambula - is as well "market orientation", like in any other agrement. Market orientation is today mainly read as "export orientation", which, in the end, means specialisation and therefore industrialization of agriculture. Small scale farmers in agro-biodiverse systems are ruled out by this mechanisms. There is - so far - no compatibility. The compatibility would rise, if (negative) externalities would be integrated in the price of goods - nitrogen surpluses for example. Only then the market could really play. But this isn't the case today. But even then - the industry logic of specialisation is against the diversity principle of nature - and therefore agriculture. To preserve (bio-)diversity we need a different orientation than market orientation. I opt for "development orientation" instead. - Development of soil fertility (i.e. carbon storage), development of sovereign seed breeding, development of local markets and resilient communities. The food system is from its nature of local character - short ways, freshness, trust, diversity. That doesn't mean no markets or import and export. But only complementing the local oriented food systems. Indias approach for food security in the Bali package makes it obvious - for food security purposes we need a different rule setting, than in other cases. - I think we have to make a decision: we should accept, that the food system is specifically different to other sectors. It needs an own set of rules, as already addressed with an AoA. But this AoA needs a different inlay than of today. As agriculture is the main stumbling bloc in the WTO and many ather FTAs - if we find a solution for this sector, many other market access negotiations could start flourishing. The rule for the food system would be something like "local is first choice". It would be in good tradition of liberal thinking, that people can decide for themselves what they want - make preferential decisions; this should be "allowed". Tarrifs and other NTBs wouldn't be necessary, so no "protection" would occur to hinder the free flow of other goods where needed.