There cannot be any unitary sentence-meaning

Within the section on the sentence-meaning of his Nyāyamañjarī (NM 5, Mysore pp. 69-onwards), Jayanta lets an opponent of sentence-holism speak. He denies the existence of a sentence-referent in the external world AND in one's cognition of it.

About that (i.e., the sentence meaning), some, to begin with, say: A sentence meaning which is ultimately real does not exist externally. It would be either separated from the word-meanings or unseparated from them. It is not separated, because no difference is seized (between the word-meanings and the sentence-meaning, which only seems to be a collection of them). In the group of words ``Bring the white cow!" the sentence meaning is only the meaning of a word denoting a quality, one denoting a substance and one denoting an action, either [taken] together or one by one. Not one by one, because the sentence meaning is not understood in this way.

There is no togetherness of them. For this would exist either ontologically or cognitively. No togetherness of the endless group of words can exist ontologically. Hence, no sentence-meaning determined by [the words featuring in a specific sentence] could be understood. Cognitively, by contrast, the togetherness is not possible (ghaṭ-), because at the time of the cognition of one word-meaning it is impossible to cognize another word-meaning, since cognitions do not [occur] at the same time.
And the phonemes are the tools for the apprehension of the word-meaning. [And] they also do not occur at the same time. How could the togetherness be made within a cognition?

atraike tāvad āhuḥ | vākyārtho nāma pāramārthiko bahir nāsty eva | sa hi padārthebhyo vyatirikto vā syād avyatirikto vā | na vyatiriktaḥ bhedānupalambhāt | gauḥ śuklā ānīyatām ity atra padagrāme jātiguṇakriyādipadārtha eva vākyārthaḥ pratyekaṃ vā syāt sāmastyena vā | na pratyekaṃ tathānavagamān na hi gaur iti padārtha eva vākyārtho bhavati |

sāmastyaṃ tu na teṣām asti tad dhi sattayā bhavet pratītyā vā | sattayā na sāmastyam aśeṣapadārtharāśer astīti na niyataḥ kaścana vākyārtho ’vadhāryate | pratītyā tu sāmastyam aghaṭamānam ayugapadbhāvitvena jñānānām ekapadārthapratītisamaye padārthāntarapratītyasambhavāt | padārthapratītiypāyāś ca varṇās te ’pi na yugapadbhāvinaḥ kutaḥ pratītikṛtaṃ sāmsastyam |

I am not completely sure I understood his point as far as the impossibility of the ontological existence of a group of words together. The author quoted by Jayanta seems to imply that we cannot imagine that each group of words has a separate ontological existence, since there if we were to admit the ontological existence of a group of words together, this should include all possible words, and one could, hence, no longer be able to determine the meaning corresponding to the sentence one is examining. But why could not separate sentence-meaning ontologically exist? It is surely anti-economical to imagine a world made of such complex objects, but I cannot understand why the option is not even discussed.
Do readers see better than I do?

On Jayanta, see this post. On the ontological status of word-meanings, see here.